We propose, as an alternative to current face recognition paradigms, an algorithm using reweighted l₂ minimization, whose recognition rates are not only comparable to the random projection using l₁ minimization compressive sensing method of Yang et al [5], but also robust to occlusion. Through numerical experiments, reweighted l₂mirrors the l₁solution [1] even with occlusion. Moreover, we present a theoretical analysis on the convergence of the proposed l₂approach.
About the Author
Jie Liang moved to the United States from China with her family to attend UCF as a first generation undergraduate. In 2010, Jie participated in a year-long research project with Dr. Xin Li that was funded by the National Science Foundation (NSF). In summer 2011, she conducted research with Dr. Frank Garvan at the University of Florida on number theory through the Summer Research Experience for Rising Senior program. Jie will begin her graduate study in the fall of 2012.
Recommended Citation
Liang, Jie
(2012)
"An Iteratively Reweighted Least Square Implementation for Face Recognition,"
The Pegasus Review: UCF Undergraduate Research Journal: Vol. 6:
Iss.
1, Article 5.
Available at:
https://stars.library.ucf.edu/urj/vol6/iss1/5
